An approach in molecular biology to elucidate structure/function relationships of a polypeptide involves the introduction of specific mutations in cloned genes for the analysis of phenotypes (Shortle, D., J. Biol. Chem. 264, 5315-5318, 1989). This reverse-genetic approach, employing site-directed mutagenesis, has facilitated the elucidation of structure-function relationships for a large number of genes. Such methods have also been successfully used to introduce desired characteristics into gene products for use in research and its applications. In some instances, such experiments have revealed intricacies of functional organization that were not apparent from the primary sequence or expression patterns (Matthews, B., Biochemistry 26, 6885-6887, 1987).
Methods of site-directed mutagenesis have evolved rapidly since the initial description of this concept (Smith, M., Annv. Rev. Genet. 19, 423-462, 1985). A common feature of the many methods is the use of synthetic oligonucleotides (primers) carrying the desired changes in the nucleotide sequence at the site of mutagenesis. This “mutagenic” oligonucleotide is incorporated into the sequence of interest by replacing the normal sequences with the designed oligonucleotide. This is accomplished by in vitro enzymatic DNA synthesis. A second step that requires the propagation and resolution of mutant and wild-type sequences in bacteria can greatly influence the rate of mutagenesis. Recently, the use of specially selected strains of E. coli that will allow enrichment of mutant molecules has improved the efficiency of mutagenesis (Kunkel, T. A., Proc. Natl. Acad. Sci. USA 82, 480-492, 1985).
Both the efficiency and the speed of mutagenesis have been improved by the introduction of methods based on the Polymerase Chain Reaction (PCR, Saiki, R. K. et al., Science 239, 487-491, 1986). Several methods based on PCR have been described that allow the introduction of mutations into the polynucleotide of interest. See Higuchi, R. et al., Nucl. Acids Res. 16, 7351-7367 (1988); Valette, F. et al., Nucl. Acid Res. 17, 723-733 (1989); Kadowaki, H. et al., Gene 76, 161-166 (1989); Dubau, L. et al., Nucl. Acids Res. 17, 2873 (1989).
These conventional PCR-based site-directed mutagenesis methods are limited to the mutagenesis of the sequences located at the termini of the amplified sequences.
Site-directed mutagenesis was made more efficient and quick by the QuikChange™ site-directed mutagenesis kit (Cat # 200518 and 200516, Stratagene). In the QuikChange kit, mutations are introduced with two complementary oligonucleotides (primers) that contain the desired mutation sites in the center. The mutagenic primers anneal to the plasmid template and are extended with a DNA polymerase (e.g., a Pfu DNA polymerase) in a temperature cycling reaction that employs non-strand displacing extension temperatures (≦68° C.). The extension products are then digested with a selection enzyme (e.g., Dpn I) to selectively eliminate parental wild-type plasmid (e.g., methylated) and parental/mutant hybrids (e.g., hemi-methylated DNAs). Then the DNAs are transformed into a host cell (e.g., E. coli) to screen for the desired mutants.
However, when mutations are located too far apart (e.g., >10 bases) to be included in one mutagenic primer pair, one must perform sequential rounds of mutagenesis using a different primer set each round. Time-consuming transformation and screening steps are required before the DNA template is available for the next round of mutagenesis. Furthermore, it is often necessary to sequence the isolated recombinant clones to identify the desired mutants, which substantially increases the time between consecutive rounds of mutagenesis.
A modification of the QuikChange method was described that allows mutagenesis of 2-3 sites (˜4 hours per site) in a single day (Kim et al., Biotechniques, 2000, 28:196-198). In this procedure, multi-site mutations are produced by carrying out in vitro dam-methylation between successive rounds of QuikChange mutagenesis. Only one transformation and DNA preparation step is required. Although high mutation frequencies were achieved (89.0% for 2-site mutagenesis; 83.8% for 3-site mutagenesis), this method is extremely labor intensive and requires gel isolation of Dpn I-resistant DNA at each round.
A method for introducing multiple site-directed mutations was recently described by Sawano et al (2000, Nucleic Acids Research, 28: e28). In the Sawano procedure, point mutations are introduced at several sites simultaneously by annealing mutagenic primers to the same strand of plasmid DNA. Unlike the standard QuikChange method, only one primer is required per mutation site, and the primer contains a 5′ phosphate. The mutagenic primers are extended with Pfu DNA polymerase and ligated using Taq DNA ligase (step 1). The reaction products are then digested with Dpn I to eliminate methylated parental plasmid DNA (step 2). Finally, double-stranded plasmid DNA is prepared by priming circular single-stranded molecules (mutant DNAs) with endogenous Dpn I fragments or an exogenous oligonucleotide, and performing 2 additional rounds of temperature cycling (step 3; reaction uses Pfu, DNA ligase, dNTPs, and NAD (nicotinamide adenine dinucleotide) carried over from the PCR reaction). This procedure was used to prepare a GFP double mutant (Y66W/T203Y; 76% efficiency) and a quadruple mutant (>70% efficiency when step 3 was carried out with T7 primer) (Sawano, supra). The average number of colonies recovered from 2-site mutagenesis was reported to be 48 cfus (30-72 cfus per experiment).
The Sawano procedure offers several advantages for producing multiple mutations. Since point mutations are incorporated at multiple sites simultaneously within one cycling reaction, the total time required to construct and analyze mutants is reduced. Moreover, unlike the standard QuikChange method, only one primer is required per mutation site. This not only represents a cost saving, but the use of one primer also raises the possibility of creating large insertions or random site-directed mutant libraries using degenerate primers. Typically, the last step in a protein engineering or directed evolution project is to carry out saturation mutagenesis, whereby all 20 amino acid side chains are introduced at one or more site(s) known to confer the desired phenotype (Miyazaki and Arnold., J. Mol. Evol. 49: 716-720, 1999). Site-specific random mutant libraries are then screened to identify the amino acid or combination of amino acids that provides the greatest improvement in activity. The Sawano procedure was used successfully with one degenerate primer to randomly mutate amino acid T203 of GFP. In this study, mutants containing 13 different amino acid side chains at residue 203 were identified among the 62 clones isolated.
There is a need in the art for a more efficient multi-site directed mutagenesis. There is also a need for a multi-site directed mutagenesis method that generates more transformants so that large numbers of random mutants can be screened.
Directed evolution methods use the process of natural selection to combinatorially evolve enzymes, proteins, or even entire metabolic pathways with improved properties. These methods typically begin with the infusion of diversity into a small set of parent nucleotide sequences through DNA recombination and/or mutagenesis. The resulting combinatorial DNA library then is subjected to a high-throughput selection or screening procedure, and the best variants are isolated for another round of recombination or mutagenesis. The cycles of recombination/mutagenesis, screening, and isolation continue until a protein or enzyme with the desired level of improvement is found. In the last few years success stories of directed evolution have been reported (Petrounia, I. P. & Arnold, F. H. (2000) Curr. Opin. Biotechnol. 11, 325330), ranging from many-fold improvements in industrial enzyme activity and thermostability (Schmidt-Dannert, C. & Arnold, F. H. (1999) Trends Biotechnol. 17, 135136) to the design of vaccines (Patten, P. A., Howard, R. J. & Stemmer, W. P. C. (1997) Curr. Opin. Biotechnol. 8, 724733) and viral vectors for gene delivery (Powell, S. K., Kaloss, M. A., Pinskstaff, A., McKee, R., Burimski, I., Pensiero, M., Otto, E., Stemmer, W. P. & Soong, N. W. (2000) Nat. Biotechnol. 18, 12791282.
DNA shuffling (Stemmer, W. P. C. (1994) Proc. Natl. Acad. Sci. USA 91, 1074710751), along with its variants (Coco et al (2001) Nature Biotechnology 19:354; Moore et al. (2001) Proc Natl Acad Sci USA. 98:3226-31; Whalen et al. (2001) Curr Opin Mol Ther. 3:31-6), is one of the earliest and most commonly used DNA recombination protocols. It consists of random fragmentation of parent nucleotide sequences with DNase I and subsequent fragment reassembly through primerless PCR. Library diversity is generated during reassembly when two fragments originating from different parent sequences anneal and subsequently extend. This gives rise to a crossover, the junction point in a reassembled sequence where a template switch takes place from one parent sequence to another.
DNA shuffling techniques are also disclosed in U.S. Pat. Nos. 6,180,406; 6,132,970; 5,965,408; 6,165,793, 6,117,679; publications WO01/29211 and WO/0129212, all of which incorporated by references.
The key advantage of DNA shuffling is that many parent sequences can be recombined simultaneously (i.e., family DNA shuffling; Crameri, A., Raillard, S., Bermudez, E. & Stemmer, W. P. C. (1998) Nature (London) 391, 288291), generating multiple crossovers per reassembled sequence.
In the Stemmer (“sexual PCR” gene shuffling, e.g., in U.S. Pat. Nos. 6,180,406; 6,165,793; 6,132,970) method, the pool of parental genes is digested with DNase I to generate random, double-stranded DNA fragments. The fragments are size fractionated to select the smallest fragments (<50 bases), thereby maximizing the probability of multiple recombination events (and increasing diversity). The fragments are randomly assembled by cross-priming during a PCR reaction carried out in the absence of exogenous primers. Finally, the diversified products are PCR amplified using terminal primers.
WO01/29212 publication discloses a different method for DNA shuffling. In this method, single-stranded DNA, generated from a pool of parental genes, is digested with DNase I and then size fractionated. The fragments are then assembled by annealing to a “scaffold” with the following properties: 1) its DNA sequence is related—but not identical to those used to prepare DNase fragments (eliminates bias due to hybridization of fragments to their own parent); 2) it is single-stranded; 3) it is prepared with deoxyuracil to allow selective removal later in the procedure. After the fragments anneal to the scaffold (in the absence of polymerase), the duplexes are treated with Taq DNA polymerase (to trim 5′ flaps), followed by Pfu DNA polymerase (to fill in gaps) and Taq DNA ligase (to ligate fragments together). The duplexes are then treated with uracil DNA glycosylase to selectively eliminate the scaffold, and the diversified products are then amplified by PCR.
There is a need in the art for generating a large number of recombinant DNA and subsequent transformants from DNA shuffling so that the chance of screening for a polypeptide product with a desired activity can be increased. There is also a need for a quicker and a simplified method for conducting DNA shuffling.